1. Field of the Invention
Embodiments of the invention relate to a compact Mid-Infrared (MIR) laser device which is tunable utilizing an external cavity. The laser device finds applications in many fields such as, molecular detection and imaging instruments for use in medical diagnostics, pollution monitoring, leak detection, analytical instruments, homeland security and industrial process control. Embodiments of the invention are also applicable to the detection of molecules found in human breath; such molecules correlate to existing health problems such as asthma, kidney disorders and renal failure.
2. Description of Related Art
MIR lasers of interest herein may be defined as lasers having a laser output wavelength in the range of approximately 3-12 μ/m (3333−833 cm−1). More broadly, however, “MIR” may be defined as wavelengths within a range of 3-30 μm. The far-IR is generally considered 30-300 μm, whereas the near IR is generally considered 0.8 to 3.0 μm. Such lasers are particularly advantageous for use in absorption spectroscopy applications since many gases of interest have their fundamental vibrational modes in the mid-infrared and thus present strong, unique absorption signatures within the MIR range.
Various proposed applications of MIR lasers have been demonstrated in laboratories on bench top apparatuses. Actual application of MIR lasers has been more limited and hampered by bulky size and cost of these devices.
One laser gain medium particularly useful for MIR lasers is found in the quantum cascade laser (QCL). Such lasers are commercially available and are advantageous in that they have a relatively high output intensity and may be fabricated to provide wavelength outputs throughout the MIR spectrum. QCL have been shown to operate between 3.44 and 84 μm and commercial QCL are available having wavelengths in the range of 5 to 11 μm. The QCL utilized two different semiconductor materials such as InGaAs and AlInAs (grown on an InP or GaSb substrate for example) to form a series of potential wells and barriers for electron transitions. The thickness of these wells/barriers determines the wavelength characteristic of the laser. Fabricating QCL devices of different thickness enables production of MIR laser having different output frequencies. Fine tuning of the QCL wavelength may be achieved by controlling the temperature of the active layer, such as by changing the DC bias current. Such temperature tuning is relatively narrow and may be used to vary the wavelength by approximately 0.27 nm/Kelvin which is typically less than 1% of the of peak emission wavelength.
The QCL, sometimes referred to as Type I Cascade Laser or Quantum Cascade Laser, may be defined as a unipolar semiconductor laser based on intersubband transitions in quantum wells. The QCL, invented in 1994, introduced the concept of “recycling” each electron to produce more than one photon per electron. This reduction in drive current and reduction in ohmic heating is accomplished by stacking up multiple “diode” regions in the growth direction. In the case of the QCL, the “diode” has been replaced by a conduction band quantum well. Electrons are injected into the upper quantum well state and collected from the lower state using a superlattice structure. The upper and lower states are both within the conduction band. Replacing the diode with a single-carrier quantum well system means that the generated photon energy is no longer tied to the material bandgap. This removes the requirement for exotic new materials for each wavelength, and also removes Auger recombination as a problem issue in the active region. The superlattice and quantum well can be designed to provide lasing at almost any. photon energy that is sufficiently below the conduction band quantum well barrier.
Another type of Cascade Laser is the Interband Cascade Laser (ICL) invented in 1997. The ICL, sometimes referred to as a Type II QCL (Cascade Laser), uses a conduction-band to valence-band transition as in the traditional diode laser, but takes full advantage of the QCL “recycling” concept. Shorter wavelengths are achievable with the ICL than with QCL since the transition energy is not limited to the depth of a single-band quantum well. Thus, the conduction band to valance band transitions of the Type II QCLs provide higher energy transitions than the intra-conduction band transitions of the Type I QCLs. Typical wavelengths available with the Type II QCL are in the range of 3-4.5 μm, while the wavelengths for the Type I QCLs generally fall within the range of 5-20 μm. While Type II QCLs have demonstrated room temperature CW operation between 3.3 and 4.2 μm, they are still limited by Auger recombination. Clever bandgap engineering has substantially reduced the recombination rates by removing the combinations of initial and final states required for an Auger transition, but dramatic increases are still seen with active region temperature. It is expected that over time improvements will be made to the ICL in order to achieve the desired operating temperature range and level of reliability.
For purposes of the present invention, QCL and ICL may be referred to under the generic terminology of a “quantum cascade laser” or “quantum cascade laser device”. The laser gain medium referred to herein thus generally refers to a quantum cascade laser in the context of the fixed wavelength embodiments. When the quantum cascade laser is utilized in a tunable external cavity arrangement as described in other embodiments herein, one of the mirror facets of the quantum cascade laser, which in the fixed wavelength embodiments serves as a partially reflecting mirror, is replaced with an anti-reflective coating so that the laser light is passed to the external cavity and impinges upon a wavelength dependent filter. This wavelength dependent filter is used to feed back to the laser gain medium a narrow band wavelength which is then preferentially amplified in the laser gain medium. In this manner, the laser output may be tuned to a desired wavelength within a range around the nominal center wavelength of the quantum cascade laser. Thus, when the quantum cascade laser is used in an external cavity arrangement, it is more accurate to refer to the lasing device as a quantum cascade laser gain medium or simply as the laser gain medium since the external cavity and not the facet mirror of the laser chip itself dictates what wavelength will experience the most gain and thus dominate the laser output.
For the purposes of the present invention, the term “subband” refers to a plurality of quantum-confined states in nano-structures which are characterized by the same main quantum number. In a conventional quantum-well, the subband is formed by each sort of confined carriers by variation of the momentum for motion in an unconfined direction with no change of the quantum number describing the motion in the confined direction. Certainly, all states within the subband belong to one energy band of the solid: conduction band or valence band.
For the purposes of the present invention, the term “nano-structure” refers to semiconductor (solid-state) electronic structures including objects with characteristic size of the nanometer (10−9) scale. This scale is convenient to deal with quantum wells, wires and dots containing many real atoms or atomic planes inside, but being still in the size range that should be treated in terms of the quantum mechanics.
For the purposes of the present invention term “unipolar device” refers to devices having layers of the same conductivity type, and, therefore, devices in which no p-n junctions are a necessary component.
The development of small MIR laser devices has been hampered by the need to cryogenically cool the MIR lasers (utilizing, for example, a large liquid nitrogen supply) and by the relatively large size of such devices hampering their portability and facility of use and thus limiting their applicability.
A fixed wavelength MIR laser device is described in co-pending application Ser. No. 11/154,264, filed Jun. 15, 2005, and incorporated herein by reference. The discussion of the prior application is made for convenience in the Background section of the specification and no admission of such disclosure as being prior art is made thereby. In the fixed wavelength device, the laser is tunable to a small degree by change in temperature of the laser gain medium either by external temperature control or by variation of the input current to the quantum cascade laser. Thus, while such laser devices are generally referred to as “fixed” wavelength, it is understood by those skilled in the art that a relatively small variation of wavelength is nevertheless available, typically less than 1% of the peak wavelength, by means of temperature control.
As disclosed in the co-pending application, FIG. 1A shows FIGS. 1A-1C show perspective views of a MIR laser device 2. FIG. 1A shows the MIR laser device 2 with the housing 4 including the lid or top cover plate 4a and mounting flanges 4b. FIGS. 1B and 1C show the MIR laser device 2 with the lid 4a removed, thus exposing the interior components. FIGS. 2A and 2B show exploded perspective, views of the various components of the MIR laser. FIGS. 3 and 4A show plan and side views respectively of the laser device and FIG. 4B shows an enlarged portion of FIG. 4A.
As may be seen from these figures, the MIR laser device is seen to include a laser gain medium 6 mounted on a high thermal conductivity sub-mount 8. There is further provided a temperature sensor 10, a lens holder 12, output lens mount 13, output lens 14, and window 16. An output aperture 18a is provided in the side of the housing 4 with the window positioned therein. The MIR laser device is also comprised a heat spreader 20, cooler 22 and electronics sub-assembly 24. The heat spreader 20 also serves as the optical platform to which the key optical elements of the laser device are secured. Thus, more precisely, element 20 may be referred to as the heat spreader/optical platform and this composite term is sometimes used herein. However, for simplicity, element 20 may be referred to as a “heat spreader” when the heat transfer function is of interest and as an “optical platform” when the platform features are of interest. The housing 4 is also provided with an RF input port 26 and a plurality of I/O leads 28 which connect to the electronic sub-assembly 24 and temperature sensor 10. These leads may extend out of one or both sides of the housing.
The output lens mount 13, especially as seen in FIGS. 2A and 2B, is seen to comprise a U-shaped support 13a, a retention cap 13b, top screws 13c and front screws 13d. The lens 14 is secured within the lens holder 12 as for example by means of glue or solder. The lens holder 12 in turn is secured within the output lens mount 13 and specifically between the lens U-shaped support 13a and the retention cap 13b. Spring fingers 13e secured to the retention cap 13b make pressure contact with the top portions of the lens holder 12 when the top screws 13c, which are threaded, are tightened down into mating threaded holes in U-shaped support 13a, to secure the retention cap 13b to the U-shaped support 13a. The front screws 13d secure the U-shaped support 13a to the optical platform 20 via threaded portions 13f screwed into threaded holes 13k within the front surface 20c of the heat spreader/optical platform 20. In this manner, the output lens mount 13, (and consequently the lens 14 itself) is rigidly and fixedly secured to the optical platform 20.
The laser gain medium 6 is preferably a quantum cascade laser, either QCL or ICL) which has the advantages providing tunable MIR wavelengths with a small size and relatively high output intensity. Examples of such a laser include 3.7 μm and 9.0 μm laser manufactured by Maxion. These quantum cascade lasers have partially reflecting and fully reflecting mirrors formed by the end facets of the laser gain material. The laser gain medium 6 typically has a size of 2 mm×0.5 mm×90 microns and is mounted directly to the high thermal conductivity submount 8 utilizing an adhesive or weld or other suitable method of securing same. The high thermal conductivity sub-mount 8 is preferably made of industrial grade diamond and may have representative dimensions of 2 mm high×2 mm wide×0.5 mm long (length along the beam path). An alternative dimension may be 8 mm high×4 mm wide by 2 mm long. Other materials may also be used as long as they have a sufficiently high thermal conductivity sufficient to conduct heat from the laser gain medium 6 to the larger heat spreader 20. The thermal conductivity is preferably in the range of 500-2000 W/mK and preferably in the range of approximately 1500-2000 W/mK. In alternative embodiments, the high thermal conductivity submount 8 may be made of a layer of diamond mounted on top of a substrate of another high thermal conductive material such as Cu or CuW. For example, the overall dimensions of the submount may be 8 mm high×4 mm wide×2 mm long (length along the beam path), and it may be composed of a diamond portion of a size 0.5 mm high×2 mm wide×2 mm long with the remaining portion having a size of 7.5 mm high×2 mm wide×2 mm long and composed, for example, of Cu or CuW. In a most preferred embodiment, the size of the housing is 3 cm (height)×4 cm (width)×6 cm (length) where the length is taken along the direction of beam propagation (optical axis) and includes the two mounting flanges 4b on each end of the housing 4.
The heat spreader 20 may be fabricated from copper-tungsten or other material having a sufficiently high thermal conductivity to effectively spread out the heat received from the high thermal conductivity sub-mount 8. Moreover heat spreader may be composed of a multilayer structure of high thermal conductivity. The high thermal conductivity sub-mount 8 may be secured to the heat spreader 20 by means of epoxy, solder, or laser welded.
The heat spreader 20 is placed in direct thermal contact with the cooler 22 which may take the form of a thermo-electric cooler (TEC) which provides cooling based on, for example, the Peltier effect. The TEC may also be fabricated from thermionic coolers or microcoolers, made from, for example, silicon germanium. As best seen in FIG. 4, the cooler 22 is placed in direct thermal contact with the bottom wall of the housing 4 and transfers heat thereto. The bottom surface of the heat spreader 20 may be secured to the top surface of the cooler 22 by means of epoxy, welding, solder or other suitable means. Alternatively, an intermediate plate may be attached between the top surface of the cooler 22 and the bottom surface of the heat spreader 20 in order to provide further rigidity for the optical platform function of the heat spreader 20. This intermediate plate may serve as a substrate on which the heat spreader is mounted. If the intermediate plate is not utilized, then the top surface of the TEC heat cooler 22 serves as the substrate for mounting the heat spreader 20.
The laser device 2 may have its housing mounted to a heat sink (not shown) inside a larger housing (not shown) which may also contain additional equipment including cooling fans and vents to further remove the heat generated by the operation of the laser.
The cooler 22 is driven in response to the temperature sensor 10. The cooler may be driven to effect cooling or heating depending on the polarity of the drive current thereto. Currents up to 10-A may be required to achieve temperature stability in CW operation, with less required in pulsed operation. Temperature variations may be used to effect a relatively small wavelength tuning range on the order 1% or less.
The lens 14 may comprise an aspherical lens with a diameter approximately equal to or less than 10 mm and preferably approximately equal to or less than 5 mm. Thus, the focal length may be one of approximately 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 mm and any fractional values thereof. The focal length of the lens 14 is fabricated to be approximately ½ the size of the diameter. Thus, 10 mm diameter lens will have a focal length of approximately 5 mm, and a 5 mm diameter lens will have a focal length of approximately 2.5 mm. In practice, the lens focal length is larger than ½ the diameter as discussed below in connection with the numeric aperture. The lens 14 serves as a collimating lens and is thus positioned a distance from the laser gain medium 6 equal to its focal length. The collimating lens serves to capture the divergent light from the laser gain medium and form a collimated beam to pass through the window 16 to outside the housing 4. The diameter of the lens is selected to achieve a desired small sized and to be able to capture the light from the laser gain medium which has a spot size of approximately 4 μm×8 μm.
The lens 14 may comprise materials selected from the group of Ge, ZnSe, ZnS Si, CaF, BaF or chalcogenide glass. However, other materials may also be utilized. The lens may be made using a diamond turning or molding technique. The lens is designed to have a relatively large numerical aperture (NA) of approximately of 0.6. Preferably the NA is 0.6 or larger. More preferably, the NA is approximately 0.7. Most preferably, the NA is approximately 0.8 or greater. To first order the NA may be approximated by the lens diameter divided by twice the focal length. Thus, selecting a lens diameter of 5 mm with a NA of 0.8 means that the focal length would be approximately 3.1 mm. The lens 14 has an aspheric design so as to achieve diffraction limited performance within the laser cavity. The diffraction limited performance and ray tracing within the cavity permits selection of lens final parameters dependent on the choice of lens material.
The small focal length of the lens is important in order to realize a small overall footprint of the laser device 2. Other factors contributing to the small footprint include the monolithic design of the various elements, particularly as related to the positioning of the optical components and the ability to efficiently remove the large amount of heat from the QCL serving as the laser gain medium 6.
The monolithic advantages of the described embodiments result from utilizing the heat spreader/optical platform 20 as an optical platform. The output lens 14 and laser gain medium 6 are held in a secured, fixed and rigid relationship to one another by virtue of being fixed to the optical platform 20. Moreover, the electronic subassembly is also fixed to the optical platform 20 so that all of the critical components within the housing are rigidly and fixedly held together in a stable manner so as to maintain their relative positions with respect to one another. Even the cooler 22 is fixed to the same optical platform 20. Since the cooler 22 takes the form of a thermoelectric cooler having a rigid top plate mounted to the underside of the optical platform 20, the optical platform 20 thereby gains further rigidity and stability. The thermoelectric cooler top plate is moreover of approximately the same size as the bottom surface of the heat spreader/optical platform 20 thus distributing the heat over the entire top surface of the cooler 22 and simultaneously maximizing the support for the optical platform 20.
The heat spreader/optical platform 20 is seen to comprise a side 20a, a top surface 20b, a front surface 20c, a step 20d, a recess 20e and bridge portion 20f and a heat distributing portion 20g. The electronic subassembly 24 is secured to the top surface 20b. The laser gain medium 6 may be directly secured to the bridge portion 20f. If an intermediate high thermal conductivity submount 8 is used between the laser gain medium 6 and the bridge portion 20f, the submount 8 is directly mounted to the bridge portion 20f and the laser gain medium 6 is secured to the submount 8. The output lens mount is secured to the front surface of the optical platform 20 via the front screws 13d. As best seen in FIG. 4A, a portion of the lens holder 12 is received within the recess 20e. It may further be seen that the surface of the lens 14 proximate the laser gain medium 6 is also contained within the recess 20e. Such an arrangement permits the lens, with its extremely short focal length, to be positioned a distance away from the laser gain medium 6 equal to its focal length so that the lens 14 may serve as a collimating lens. The remaining portions of the lens 14 and the lens holder 12 not received within the recess 20e are positioned over the top surface of the step 20d. The heat distributing surface 20g of the heat spreader/optical platform 20 is seen to comprise a flat rigid plate that extends substantially over the entire upper surface of the thermo electric cooler 22. Other than the screw attachments, the elements such as the temperature sensor 10, laser gain medium 6, high thermal conductivity submount 8 and electronics subassembly 24 may be mounted to the heat spreader/optical platform 20 by means of solder, welding, epoxy, glue or other suitable means. The heat spreader/optical platform 20 is preferably made from a single, integral piece of high thermal conductivity material such as a Cu or CuW alloy as non-limiting examples.
The housing 4 is hermetically sealed and for this purpose the lid 4a may incorporate an “O” ring or other suitable sealing component and may be secured to the housing side walls in an air tight manner, e.g., weld or solder. Prior to sealing or closure, a nitrogen or an air/nitrogen mixture is placed in the housing to keep out moisture and humidity. The window 16 and RF input port 26 present air tight seals.
The temperature sensor 10 may comprise an encapsulated integrated circuit with a thermistor as the temperature sensor active component. A suitable such sensor is model AD 590 from Analog Devices. The temperature sensor 10 is positioned on the heat spreader 20 immediately adjacent the laser gain medium 6 and is effective to measure the temperature of the laser gain medium 6. As best seen in FIGS. 1C and 2A the temperature sensor 10 as well as the laser gain medium 6 are in direct thermal contact with the heat spreader 20. The temperature sensor 10 is in direct physical and thermal contact with the heat spreader 20. In one embodiment, the laser gain medium 6 is in direct physical and thermal contact with the high thermal conductivity submount 8. However, in other embodiments, the high thermal conductivity submount 8 may be eliminated and the laser gain medium 6 may be secured in direct physical and thermal contact with the heat spreader 20 with all other elements of the laser device remaining the same. The temperature sensor 10 is connected to the I/O leads 28. The temperature output is used to control the temperature of the cooler 22 so as to maintain the desired level of heat removal from the laser gain medium 6. It may also be used to regulate and control the injection current to the laser gain medium 6 which also provides a temperature adjustment mechanism. Varying the temperature of the laser gain medium 6 serves to tune the laser, e.g., vary the output wavelength.
The electronic sub-assembly 24 is used to control the laser gain medium 6 by controlling the electron injection current. This control is done by using a constant current source. In effect the quantum cascade laser behaves like a diode and exhibits a typical diode I-V response curve. For example, at and above the threshold current, the output voltage is clamped to about 9 volts.
FIG. 5 shows a schematic diagram of the electronics subassembly 24. The electronics subassembly is seen to comprise capacitors C1 and C2, resistor R1, inductor L1, a summing node 30, switch 32, and leads 28a and 28b. A trace or transmission line 34a, 34b (see also FIG. 3) interconnects components. The polarities of the electronics subassembly 24 are selected for a chip arrangement in which the epitaxial layer of the quantum cascade laser is positioned downwardly. Polarities would be reversed if the epitaxial layer side is positioned upwardly.
The RF input port 26 is seen to be fed along the transmission line 34a to one side of the first capacitor C1. Resistor R1, which may comprise a thin film resistor, is positioned between capacitors C1 and C2 and connects the junction of these capacitors to ground. The capacitors and resistor implement an impedance matching circuit to match the low impedance of the quantum cascade laser with the 50 ohm input impedance line of the RF input cable. Transmission line 34b interconnects inductor L1 with the switch 32 and connects to the laser gain medium 6. The inductor L1 is fed by a constant current source (not shown) via one of the I/O leads, here identified as lead 28a. Inductor L1 serves to block the RF from conducting out of the housing through the current lead 28a. Similarly, a function of the capacitor C2 is to prevent the DC constant current form exiting the housing via the RF port 26. The switch 32 may take the form of a MOSFET and is biased by a switching control signal (TTL logic) fed to I/O lead 28b. Controlling the duty cycle of this switching control signal controls the relative on/off time of the MOSFET which is operative to pass the drive current either to the laser gain medium 6 (when the MOSFET is off) or to shunt the drive current to ground (when the MOSFET is on). With TTL logic in the illustrated circuit, a 0 volt switching control signal turns MOSFET off and thus the quantum cascade laser on, and a −5 volt switching control signal turns the MOSFET on and thus the quantum cascade laser off. By controlling the switching control signal duty cycle, pulse or cw operation may be realized.
An RF input signal is fed to the RF input port 26. This RF signal is used to frequency modulate the drive current signal to the laser gain medium 6 and is summed with the drive current at the summing node 30. Frequency modulation is commonly used to improve sensitivity in absorption spectroscopy. The center frequency is scanned across the expected resonance (using, for example, temperature tuning achieved by variation of the TEC cooler 22 or variation of the current fed to the quantum cascade laser). Frequency modulation places sidebands about the center frequency, and during the wavelength scanning a strong RF modulation may be observed when off resonance due to an imbalance in the absorption of the frequency sidebands. FM modulation thus effectively produces an AM modulation of the absorption signal. However, at resonance, the effect of the frequency sidebands is of opposite phase and equal magnitude so they cancel out. Sweeping the frequency about the resonance peak (dip) using FM modulation thus permits one to pinpoint more accurately the center of the absorption line which corresponds to a minimum in the AM modulation over the sweep range. Techniques for FM modulation are well known to those skilled in the art and reference is made to the following articles incorporated herein by reference: Transient Laser Frequency Modulation Spectroscopy by Hall and North, Annu Rev. Phys. Chem. 2000 51:243-74.
The quantum cascade lasers utilized herein have an intrinsically high speed. Thus, to effectively perform FM modulation, the modulated signal must be injected in close proximity to the quantum cascade laser to eliminate any excess inductance or capacitances associated with the laser connections to the RF signal. This is especially important in quantum cascade lasers which present a fairly low impedance and thus the reactance of the connections will critically limit the speed with which the device can be modulated. The circuit design as disclosed herein presents an extremely small footprint for connections of the RF input to the quantum cascade laser. Thus, for a 1 GHz modulation frequency, a representative range of transmission lengths from the RF input port 26 to the laser gain medium (QCL) (the sum of 34a and 34b) is 2-4 cm or less generally less than or equal to 4 cm. A preferred value is approximately 3 cm. If one desires to choose a broadband input for the FM modulation restricting the maximum frequency to 1 GHz, then the optimal transmission length is approximately 1 cm or greater. Such a transmission length would permit operating at 100 MHz for example or other values up to the 1 GHz level. Thus, in performing FM modulation of the quantum cascade laser a small transmission path is optimal in order to present a low inductance path to the QCL thereby permitting relatively high modulating frequency to be used. The small transmission paths may be suitably contained with the structures of the disclosed electronic subassembly 24.
It is noted that the entire electronic subassembly 24 is rigidly and fixedly mounted on the heat spreader 20 which serves, as indicated above as an optical platform. The fixing of the transmission lines and other electronic components to the optical platform achieves a rugged design which is largely insensitive to outside vibrational disturbances.
The input leads 28 are seen to comprise leads 28a and 28b and the RF input port 26 described above. Other I/O leads to the housing 4 include the + temp drive signal lead for the TEC to cause the TEC to be heated, a − temp drive signal lead to cause the TEC to be cooled, the temperature sensor input lead to provide a bias voltage to the thermistor temperature sensor, a temperature output lead to provided an output signal for the temperature sensor and a ground return path for the constant current input to the quantum cascade laser.